Metaphor and Psychoanalysis: Metaphor and the Violent Act

by Donald Campbell , Henrik Enckell

August 12, 2006


During the treatment of violent individuals who were, incidentally, highly verbal, the authors noticed that physical assaults were often preceded by the perpetrator’s use of metaphors. It was observed that the linguistic metaphors failed to function as ordinary ‘as if’ devices and became ‘concretised’. When this occurred, the perpetrators resorted to a physical attack. In this paper, the authors argue that the capacity to interconnect (which is considered to be the essence of psychic work) is dependent upon what can be conceptualised as a primary mental ‘frame’ or ‘warp’. Distortion of the warp will, in turn, weaken the ‘weaving’, or interconnecting function of the ego, which is considered analogous to the interconnecting in linguistic metaphors. Clinical material from the treatment of three violent men (two in psychotherapy and one in analysis) is used to illustrate the hypothesis that the concretised use of metaphor represents a restitutive, but failed attempt to maintain a psychic coherence in the face of an imminent breakdown.


This article originally appeared in the International Journal of Psychoanalysis 86 (2005): 801-823.


We refer to violence as the intended infliction of bodily harm (Walker, 1968). With some notable exceptions, e.g. Williams (1960), until recently there have been only a few reports on the psychoanalytic or psychotherapeutic treatment of actual perpetrators. However, numerous recent publications (Sohn, 1995; Fonagy and Target, 1995; Perelberg, 1995, 1999; Glasser, 1998) reflect a growing interest in the psychology of concrete and manifest violence. In this paper, we will present clinical material from the once-weekly psychotherapy of two violent men treated by one of us at the Portman Clinic2 along with material from Perelberg’s five-times-weekly analysis of a violent individual to develop some observations made in a research study into the psychodynamics of the violent act.3

     We are aware that the relatively low frequency of sessions in the psychotherapies presented makes the clinical material somewhat ‘thin’, but it has to be recognised that perpetrators seldom start a regular psychoanalysis. We consider our views as preliminary until they can be supported through the study of intensive psychoanalytic treatment of violent patients, such as the case by Perelberg (1995, 1999), which we have included.

     In the cases observed in the study, it was noted that the perpetrators—just prior to the assault—formed their experiences in what can be called ‘concretised’ metaphors. While metaphors ordinarily function as representations through which experiences can be contained, concretised metaphors are perceived as presentations stating a merciless reality; instead of being functioning symbols, concretised metaphors are experienced as immovable facts with which one cannot negotiate. In this paper, we study the links between the violent act, on the one hand, and the concretized metaphor, on the other. Through the picture offered by the concretised metaphor we hope to shed some light on actual violence. The object of study will be the concrete act, not aggression in general; through this focus we hope to shed some light on the origins of actual violence.

The matrix of experience

In this section we will give an outline of a theoretical model (based on the findings of other authors) that has helped us make sense of the clinical findings to be studied.

     According to Stern (1985), the first mental imprints are organised around satisfying experiences in mother-infant interactions. These first stable representational configurations, called ‘islands of consistency’, are the nodal points in what Stern calls the first mental activity, i.e. the interconnecting of the configurations in question. The importance of the connecting activity stems from its adherence to the first sense of a self (the ‘emergent self’); the experiential dimension of the linking is the sense of self. Stern seems to imply that the process of connecting is the basis for all later ‘senses of the self’.

     A psychical interconnecting thus becomes the condition for the experiential self. What is the condition for the interconnecting? Using Shaftesbury’s classical metaphor, one might ask: what is the warp of the weaving mind? Two concepts have helped us elaborate a picture of this structure.

     Gaddini (1982), Green (1986) and Salonen (1989) have studied ‘primary identification’ to grasp how the infant gains a basic psychic matrix. According to the definition of the term (Freud, 1921), the subject is constituted in an identification with the caring adult, and this primary identification is the condition for object relations proper. Gaddini, Green and Salonen state that the primary identification not only conditions the formation of the subject, but it also makes vital connecting, and thus psychical aliveness, possible. According to these authors, the basic matrix serves as a fundamental representational ‘warp’ or ‘frame’ supplying stability to the representational world; as a conditioning structure one can perceive it only in its failures, as in the fragmentation and loss of psychic vitality in schizophrenic breakdown (Salonen, 1979).

     The other concept informing our model is the ‘body ego’. Following Lehtonen (1997) the body ego is a representational ‘screen’ capable of transforming bodily stimuli into psychic work. While Lewin (1946, 1948) coined the concept ‘dream screen’ to picture a conditioning structure for all dreams, Lehtonen sees the body ego as a ‘representational screen’ underlying mental processes. The body ego can thus be seen as a structure capable of transforming somatic sensations into psychic elaboration, and thus conditioning the origins of what Laplanche and Pontalis (1973) identify as psychic work, i.e. psychic interconnecting.

     Common to all these authors is a topographic and hierarchical conception of psychical functions. Basic functions—and structures—are not in themselves observable, but they form the basis for more elaborate ones. This general model can be found in the works of many authors, for example, Sandler and Sandler’s papers (e.g. 1983, 1984) on the ‘past’ and the ‘present’ unconscious.

     Although the body-ego matrix is a theoretical construct, we believe it can describe the infant’s ensuing faculties of bodily relaxation, of falling asleep and of gaining behavioural restfulness, for example. We also believe it can be described as the structural condition for primary hallucination and the construct of the wish (Opatow, 1997), at the same time as it could be said to represent the psychic structure for ‘procedural’ behavioural phenomena (Sandler and Sandler, 1998). Accordingly, the development of this body-ego matrix can be inferred from observable developmental phenomena implying basic introjections, at the same time as it can be viewed as the structure transforming bodily needs into primary psychic functions.

The theory of metaphor

Metaphor is a rhetorical figure combining words not usually combinable, thus suggesting a resemblance between domains ordinarily perceived as clearly separate (Ricoeur, 1978).

     After Sharpe’s (1940) pioneering work in the field, many psychoanalytical authors have used the theory of metaphor to investigate both theoretical and clinical matters. The application of the metaphor model has been especially marked during the last two decades (e.g. Arlow, 1979; Shengold, 1981; Kitayama, 1987; Smith, 1992; Corradi Fiumara, 1995; Ogden, 1997; Melnick, 1997; Gammelgaard, 1998; Borbely, 1998; Rizzuto, 2001), but this is not exclusive to psychoanalysis. In recent decades, the metaphor model has been applied not only in literary theory, but also in cognitive research, gender studies, epistemology and the theory of science. We cannot review here the extensive psychoanalytic literature in the field, but want to draw attention to one fact: for many psychoanalytical authors the metaphor is not only a linguistic device, but also a model describing certain general psychic processes. While a linguistic metaphor elaborates words, a psychoanalyst’s metaphor elaborates, in addition, non-verbal representational means such as perceptions, feelings, thoughts or memories. The linguistic model is thus also applied to non-verbal representational means, benefiting from the very preciseness of the model in question.

     The function of metaphors can be described two ways: the first one can be exemplified by Black’s (1962, 1979) works on models and metaphors; the second through Richards’s (1936), Beardsley’s (1958, 1962) and Ricoeur’s (1978) emphasis on the interaction between the different parts comprising the metaphor.

Black’s model

Concrete models can show in tangible form something difficult to perceive in reality. A model of a knee joint shows not only the anatomical structure, but also how it functions. Theoretical models function in a similar way, although here the material used is not tangibly concrete, but abstract. Building his theories, the scientist uses constructs that may be graphic or taken from tangible domains to visualise something that lacks conceptual tools of its own. Through the graphic or well-known material, the scientist studies something elusive or newly found. Psychoanalysis offers many examples of this, for example, Freud’s (1920) attempt to study basic psychic processes through a biological model.

     According to Black (1962, 1979), metaphors function like models. Through the use of well-known, ordinary words, the poet is able to describe and study an experiential field hitherto unseen. The use of a set of words or concepts from one field of experience can open up the domain of another more elusive one. The use of metaphors is motivated by the lack of ‘direct’ expressional means. This way of describing the function of metaphors can be supplemented by another one.

The inter-actional model

According to many authors (e.g. Richards, 1936; Beardsley, 1958, 1962; Ricoeur, 1978), a metaphor consists of a combination of words uncombinable according to ordinary logic. Proust’s description of the opera is an example. In Remembrance of things past (1934), the narrator sees an opera performance as happening in an undersea world. In this extended metaphor (simplifiable as ‘the opera stage is an undersea happening’), two clearly separate domains are brought together. The ‘metaphorical twist’ (Beardsley, 1962)—the ‘clash’ between the words or domains—leaves the reader in an awkward position of non-understanding (what does the undersea world have to do with the opera?), which, in its turn, leads to ‘semantic work’. Each word comprises many potential meanings, which could be found in the dictionary, and the totality of these meanings can be said to set up a ‘semantic field’. In the work following the metaphorical twist, the ‘horizons’ of the words’ semantic fields are drawn towards each other, resulting in a bridging of distant fields. In the example from Proust, the ways the reader understands both opera and the undersea world are extended, and gradually he comes to see resemblances between the worlds in question. The clash leads to a widening of the semantic fields of the words or domains and the effect of this is an ability to see something new; as the words in the metaphor become extended, so they become able to describe domains hitherto not illuminated by verbal means.

     It seems clear that metaphors are not just ‘rhetorical ornaments’, but investigatory instruments supplying a means to see dimensions of reality not describable in other ways. Through the combination of words belonging to different logical contexts, it is possible to view connections hitherto unseen; through the medium of one set of words, it becomes possible to view something situated in a different world of meaning.

Metaphor: A working model for psychic functioning

As stated above, many authors in psychoanalysis (e.g. Arlow, 1979; Melnick, 1997; Ogden, 1997; Borbely, 1998; Rizzuto, 2001) have seen the mind as metaphorising in a general sense. The analysis of basic psychoanalytic concepts seems to confirm a view that the psyche works through the medium of non-verbal metaphor to establish links and organise mental contents. To illustrate this, the theories of dreams and transference might serve as examples.

     According to Freud (1900), dreams are instigated by unconscious wishes aroused during the day preceding the dream. By their insistence, these wishes (allowed a relatively free existence during the night due to motoric inhibition) disturb sleep. The disturbance motivates the construction of the dream whose function it is to protect sleep.

     A simple way to escape from a disturbing wish is to have it fulfilled. In primary process logic this is equal to attaining perceptual identity, that is, to perceive the wish as fulfilled. To attain this goal, the wish-fulfilment has to be actualised. The wish is, however, situated in the unconscious, which implies a lack of representational means through which the wish could be perceptually put on stage. Therefore, the wish has to be transferred to the system preconscious.

     Preconsciousness comprises the day’s residues, namely, fresh reminiscences. In order to actualise a representation that can be perceived, the subject ‘chooses’ a representational means at hand, that is, the dreamer chooses the day’s residues, which in the chosen combination come to represent a configuration that in itself lacks perceivable representation. This idiosyncratic combination gives form to a reality that in itself is unrepresentable at a conscious level. Freud describes the incarnation of unconscious wishes on the stage of the day’s tangible residues as ‘transference’.

     As others (Lewin, 1971; Bird, 1972) have stated, transference is constructed as a dream. What Freud (1905, 1914) calls an invisible composition—analogous to what later authors conceptualise as unconscious fantasies (Klein, 1952; Arlow, 1969a, 1969b)—becomes activated in a psychoanalytical treatment. This composition lacks direct representational means, but may be actualised in the relationship with the analyst. In terms of drama, Loewald (1975) described how the analysand dramatises an unconscious manuscript. The manuscript or composition is actualised in a repetitious behavioural and relational scheme, but also in the subjective experience and perception of the analysand. The analysand constructs a vision of his analyst like the sleeper constructs a dream; if the latter uses the day’s residues, the former uses ‘perceptual residues’—everyday observations available for idiosyncratic combinations.

     Thus, an unconscious configuration (a wish, or unconscious fantasy) is transferred to a medium in which it finds representation, and, by this means, actualisation. Freud calls the mechanism—in dreams and in clinical situations—transference. Übertragung is equal to metaphor (Über-tragung/meta-phoros), although the translation into English might blur the fact that the words originally meant the same thing. The first person to realise the identical meaning of these terms was philosopher of language Richards (1936)

The collapse of the capacity to use functioning metaphors

We believe that the metaphor model can be seen as a model of general psychic work, defined as the work of representational interconnecting (Laplanche and Pontalis, 1973). Many authors (Arlow, 1979; Szajnberg, 1985; Borbely, 1998; Enckell, 2001) have stated that work in a psychoanalytic treatment can be compared to reading a metaphor. Psychic work is done through an interconnecting of representational fields, analogous to the ‘semantic’ work of reading a linguistic metaphor.

     We stated above that the function of metaphors can be illustrated by dreams and transference. In both of these phenomena it has been inferred that the instigator of the ‘metaphors’ in question (i.e. the dream and the transference) is the activation of a configuration (an unconscious wish or an unconscious fantasy) lacking representational means. We also described a model in which a primal psychic matrix is the condition for psychic linking and vitality. We suggest that this matrix can be equated with the structure of the unconscious wish (Salonen, 1979) and fantasy. This leads to the conception of a basic mental matrix—a psychic warp—which makes psychological weaving, and thus the use of metaphor, possible. As we know, the warp consists of lengthwise threads in a loom. In the weaving process, these are crossed by the filling threads, or ‘woof’. The warp is stouter than the woof, and it becomes invisible in it. Accordingly, the warp is the structure that supplies stability to the woof, and once the woof is completed warp does not show. As a conditioning structure, one can infer it only when it fails and the woof becomes unravelled.

     ‘Psychic warp’ is a necessary, but not sufficient, condition for a functioning metaphorisation. During stressful situations, psychic weaving might come to a standstill, and this happens more easily if psychic structures are incompletely developed as in borderline and psychotic conditions. In the following clinical material, we will focus on situations in which we suggest that weaving has become ‘torn’, and the warp, accordingly, laid bare. In situations like these, the structure underlying a functioning metaphorisation is manifested in a ‘direct’ way; something usually invisible becomes concretely present. This is analogous to the burning down of a house. The destruction (comparable to the tearing of the woof, i.e. the metaphorising function) unveils the iron constructions (comparable to the warp) in the walls. Later, we will examine clinical phenomena we view as vicissitudes of this uncovering, namely, so-called ‘concretised’ metaphors and the violent act. We want to stress that the picture of warp and woof is a model, and the function of models is to picture something through a ‘tangible’ medium; through our model we hope to make some clinical phenomena more understandable.

‘Concretised’ metaphors

In the psychoanalytic literature on metaphors, the theme of concretised metaphors is often evident. A vignette described by Kitayama (1987) illustrates this. A psychotic man complained about his sleeplessness. He could not fall asleep due to a continuing light. This man had called his former girlfriend ‘my sunshine’. It turned out that the thought of the girlfriend kept the patient awake.

     This can be conceptualised as a word-presentation turning into thing-presentation, but also as a concretisation of a metaphor. The girlfriend is not ‘like’ sunshine any more; she is sunshine. The ‘copula’ of the metaphor, the ‘is’, turns from being an implied ‘as if’ into an ‘is’. Segal (1957) coined the term ‘symbolic equation’, and Searles (1962)—stressing the concomitancy of the transformation from representation to presentation with a psychotic breakdown—calls it ‘desymbolization’. The phenomenon of concretised metaphors is widely accepted as a corollary of psychotic functioning. We propose that these metaphors may be viewed as restitutional efforts.

The case of Mr Wilson

Mr Wilson, an articulate 26-year-old barrister, asked for help with his violence towards his wife because he feared she would leave him and take the children with her. He remembered his alcoholic mother as withdrawn and narcissistically preoccupied. When it suited her, she could be emotionally seductive, confiding in him that they had a special understanding and closeness that she did not share with his sister or father. However, he remembered feeling repeatedly betrayed and abandoned by her when she failed to defend him against his father’s accusations and beatings.

     During a session, Mr Wilson told his analyst that his wife turned her back on him in the kitchen. ‘I was just trying to get her to listen to me. She wasn’t paying attention. She didn’t hear what I said. Then she turned around and walked away. I can’t remember what happened next; everything went black. I was hitting her.’

     Before the next session Mr Wilson was kept waiting two minutes. He began the session saying that he had noticed it was dark outside as he came up the stairs. He then stared out of the window and lapsed into a long silence which he broke to say: ‘I was looking at the blackness. I just wanted to punch the blackness’. The therapist suggested that he may have felt abandoned as he was not asked to come up to the consulting room at the time his session was scheduled to begin. Mr Wilson then told the therapist that he and his girlfriend had gone away for a weekend recently. In the middle of the first night, Mr Wilson awoke in the darkened hotel room in an acute anxiety state. He had become disoriented in time and place. His body boundaries became blurred and he felt utterly alone as though ‘floating off into outer space’. In this psychotic state, Mr Wilson had forgotten that his girlfriend was lying next to him on the double bed. He had begun to ‘punch the blackness’, in desperation, ‘to find someone, something, anything’. Mr Wilson constructed his fears and anxieties about abandonment—an experience of desolation—via a metaphor of ‘blackness’. Mr Wilson’s metaphor failed to contain his primitive fears. The metaphor of blackness lost its symbolic representation and became a concrete object that he punched.

     Mr Wilson’s violence functioned to retrieve an object that threatened his survival by abandoning him. His motor response of punching the darkness or his girlfriend when he felt threatened by physiological deterioration was a last-ditch attempt to retrieve the woman whom he believed was essential to his survival.

     When Mr Wilson was 11 years old, he had become ill and was hospitalised for five weeks. The doctor had suspected polio. His mother did not visit or communicate with him while he was in hospital.

     During his once-weekly psychotherapy it became clear that Mr Wilson experienced his mother’s absence from hospital in two ways: first, as a rejection of his masculine body and, second, at a deeper level, as being cut off from supplies which were vital to his survival: he felt like a starving infant who was left to die.

     We stated earlier that the basic condition of functioning metaphors is an intact primal matrix or, metaphorically speaking, an intact mental warp. If the weaving function fails—if the woof is torn and the warp unveiled—metaphors per se may survive, but they function in a different way. For instance, a psychotic person dreams, but his dreams are not experienced as symbols through which something may be investigated. Analogously, the transference for a psychotic person may not be experienced as a symbolic one. For the metaphors to be workable as symbolic structures, the weaving function has to be intact; the tearing of the woof leads to a ‘desymbolization’ (Searles, 1962) of structures formerly used as symbolic representations.

     A vital psyche can be characterised by mental connective activity. It can be hypothesised that a subject being threatened by a psychotic breakdown tries to use a linking means in order to restore, or keep up, a psychic activity both processing instinctual demands and keeping up an integrative function. The metaphor is a device to link different psychical fields and to bind the fields in question into an integrated whole. The condition for this linking is, however, an experience of difference between the psychic fields in question. In cases where the differentiation between subject and object fails, there follows an overall difficulty in differentiation. What until now has been different turns into sameness, and between identical representational fields no metaphor can be built. The helpful and vital connecting activity comes to a standstill.

     We suggest that the person facing a threat of psychosis keeps up the metaphoric function in order to save his integrity. However, if the necessary basis for integrative interlinking fails—as in cases of de-differentiation between self and object—the metaphoric structure is formally established but remains abortive in function. For these persons a metaphor is not a representational figure through which something is elaborated (thus making up a psychological link), but a presentation of a reality; thus the reality-testing function fails. As the different parts are identical, no connecting is possible and the weaving stops. The result is a metaphor presenting a reality. We think this can be seen in Mr Wilson’s experience of abandonment, of being cut off from necessary supplies, which is a blackness to be punched.

     We propose that a person suffering from destruction of the weaving function—as in cases of de-differentiation between self and object—may try to keep up a psychic coherence and an elaborating function by an ordinarily linking means, that is, by using the device of a metaphor. However, as the functional basis is torn, the attempt to uphold integrity by this means fails.

The violent act

In this section we attempt to draw a picture of so-called ‘self-preservative’ violence (Glasser, 1998): violent acts motivated by a need to preserve an experience of psychic integrity and narcissistic wholeness. From a subjective point of view, these acts are self-defensive. There are, naturally, other forms of violence.

     In infancy, disturbances in physiological homeostasis lead to immediate and automatic motoric reactions like screaming, clenching of fists, kicking and waving of arms. The aim of these is to restore the homeostatic balance. Gradually, these reactions develop into organised motor patterns.

     The child’s reactions do not take place in a vacuum, but in a relationship with a caregiver, that is, the motor patterns become integrated into primal object relations. The child’s concrete acts turn into goal-directed attempts to get something from the mother. Those acts which are successful become reinforced and are repeated, while those which are unsuccessful, that is, do not stimulate a positive response, are eventually abandoned. The good enough mother uses these behavioural patterns to ‘read’ her child. For the child, these acts are attempts to attain a homeostatic balance with its internal state and its external object. For the caregiver, they are signs that convey the child’s needs and feeling states. Gradually the acts attain an ‘iconic’ function for the child. We have borrowed a term from philosophy meaning a closed representation which does not refer to anything outside of itself. The child’s acts become primitive representations, or what can be called presentations of experience.

     These motor acts are, naturally, bodily. Freud (1925) stated that a ‘taking in’ is a first representation of approval, while a ‘spitting out’ becomes equal to denial. Bodily motor schemes (e.g. bowel movements) are the first mental representations of a specific kind of pleasure. This representational level is echoed in archaic hysterical phenomena, in the ‘theaters of the body’ as described by McDougall (1989), but also in ‘procedural’ behavioural schemes (Sandler and Sandler, 1998). In normal development, these primordial representations can be inferred from motoric gestures which are not just automatic reflexes, but which the child has acquired (e.g. a gesture of avoidance the child has picked up from the mother). These gestures express—directly—a certain experience, although the child might not be aware of the fact. We suggest that these behavioural acts reflect the constitution of ‘iconic’ mental configurations. Following this line of thought a motor act becomes a primordial presentation of a mental experience (a level we here call ‘iconic’) in which object (in this instance, the experience) and representation are identical. After this, schemes of bodily functions and sensations might become what Székely (1962) has called ‘archaic meaning schemata’. Archaic meaning schemata are representational schemes taken from bodily functions and are used to categorise and to give meaning to the world. To illustrate this Székely cited the toddler who exclaimed ‘Car do big!’ when he saw a parcel fall from it. Defecation had become a scheme through which this child perceived his world. While an icon is a presentation of an experience, a meaning schemata is a pattern which mediates (i.e. represents) outer or inner realities.

     Lewin (1971), Lakoff and Johnson (1980), Shahly (1987) and Melnick (1997) have drawn attention to the fact that schemes comparable to the ones Székely describes may be compared to metaphors. A domain is viewed through a configuration well known to the child; representational schemes taken from bodily sensations and functions have become configurations which help the child to give form to both internal and external worlds. Accordingly, one can see a developmental line in which the motor act is primarily an automatic response, then also an iconic presentation of an experience, turning into a representational scheme usable as a metaphor for different domains of reality.

     For the small child disturbances in the homeostatic balance are experienced as serious threats, and the responses are active and dramatic. Object-directed responses in particular might be viewed as violent. Basing his conclusions on a study of a large number of forensic psychiatry patients (predominantly homicides), Tuovinen (1973) noted that the process leading to a violent act is often initiated by an experienced threat towards the body ego, conceptualisable as a threat to the integrity of viable narcissistic wholeness. Partly due to regression, this threat is experienced as a threat against psychic integrity; sometimes as an experience of imminent destruction. The violent act is—from a subjective point of view—an adaptive attempt to regain psychic integrity.

     The process leading to the violent act might be viewed as a regression in representational functioning. The perpetrator might ordinarily be capable of symbol formation (at least to some degree), but the threat to the body ego overwhelms the ego’s capacity for symbolic representation. The end point of this regression is the representative level of an iconic equation described above, namely, a concrete presentation in action, which also configures the experience in question. Despite the concreteness, this formation saves the subject from a psychotic breakdown, giving the experience a form without which it might lead to fragmentation.

     This picture comes close to that of Fonagy and Target (1995) who see the violent act as a vicissitude of a failing mentalising capacity. When ordinary mentalising comes to a standstill, experiences are represented in bodily terms, that is, in a violent act. In this paper, we describe this breakdown through the concept of metaphor. Although metaphors are symbolic devices, there are certain differences between a metaphor and a symbol per se. A symbol is a representation with a double meaning (Ricoeur, 1970): through a manifest meaning (e.g. ‘going astray’) one comes to a latent one (i.e. being sinful). In the symbol, the two meanings are arranged like a chain, and one comes to the hidden meaning through the apparent one. In the metaphor, on the other hand, we have two apparent meanings which do not fit (e.g. ‘the opera is an undersea world’). Owing to the clash between the two manifest meanings, a semantic work has to be done, and through this a new, extended, meaning is born.

     According to our view, the symbol is a suitable model for subjects who use repression, that is, for whom a hidden meaning becomes apparent through the manifest one. The subjects here described, however, do not lean on this defence mechanism. We thus feel the mentalising capacity described in this paper is better described through the ‘weaving’ of the metaphor model.

Mr Giles’s propensity to violence

During the course of his therapy, Mr Giles reported 12 physical assaults on people, mostly punching, but occasionally choking, and mainly against women with whom he was sexually involved. The attacks were often preceded by drinking alcohol and by arguments. The initial intent was sadistic, but occasionally the assaults had a more negating or eliminating aim, as the time when Mr Giles ‘discovered’, after punching his girlfriend Sylvia, that he had his fingers in her eyes and was trying to gouge them out. After this incident, Mr Giles became frightened that he would permanently damage someone. He was sent to the Portman Clinic in London. As was the case with Mr Wilson, metaphor failed to contain psychic conflict within Mr Giles’s internal world. He described an attack: ‘We had been arguing, my girlfriend Sylvia and me, a typical slanging match, when I thought, "I can’t take her nagging, naïve arguments any longer. Before I know, they’ll be pushed up my arse". Then I punched her in the face again and again’.

     During the argument Mr Giles used a metaphor (‘pushed up my arse’) to form an experience of threatening intrusion. Usually a metaphor like this functions as a helpful device through which an experience is given shape. In these cases, it is usually clear that the expression is something through which something else—in this case, an experience of mental intrusion—is configured. For Mr Giles, this seems to have been the case, initially. However, as the argument escalated, the expressive function changed from being a representation to becoming a presentation. Mr Giles started to experience Sylvia as actually intruding, and this gave way to violence.

Mr Giles’s account of an assault

‘It started when we kissed in the dark. I wanted to get inside Sylvia’s mouth. Then I felt I was drowning, smothered. Started punching her. I got away into another room. I shouldn’t have come out of that room but the blackness suffocated me. I couldn’t breathe. I went mad. Didn’t hit, I bit her.’

     The intensity of his wish to invade Sylvia leaves Mr Giles, when in the blackness of the room kissing her, with a conviction that he has succeeded in getting inside her. Getting inside her was suffocating. He perceived her as the suffocator. The intensity of the wish—heightened by the loss of vision in the dark—converted this into a concrete, suffocating reality.

     In principle, the experience could have been contained in a metaphor of ‘drowning’. However, instead of leading to an ordinarily functioning metaphor, the wish led to fears of physiological disintegration, which were expressed via a concretised metaphor of actual intrusion and suffocation. In Segal’s terms, this is an example of a symbol formed by projective identification, with the result that the symbol does not represent the object, but is equated with the object and functions to deny the separateness between the subject and the object (1978, p. 316). The ‘reality’ of the metaphor led to a motor response. Mr Giles had to fight his way out of an engulfing object to survive. By projective identification Sylvia became engulfing and life threatening. In biting Sylvia, Mr Giles identified with the aggressor and tried to devour her.

Mr Giles’s history

Mr Giles, a single, sometimes unemployed, 30 year-old shoe salesman, had the ‘gift of the gab’ and enjoyed a sophisticated use of language. He grew up in a small house which ‘felt crowded’. He was an only child. The family was dominated by the father who was depressed, alcoholic and violent—physically attacking Mr Giles and his mother, while his mother was long suffering and seductive. Mr Giles had no early memories of his mother but he imagined her preoccupied with fear and anxiety over her violent, unpredictable husband. Mr Giles said that the worst thing was the confusion; not knowing when the next violent outburst would happen. Mr Giles was a binge drinker, melancholic and, at times, appeared schizoid in sessions.

     In the course of his psychotherapy, it became clear that Mr Giles was traumatised by a childhood illness that contributed to his violence in adulthood. When he was 8, he’d had rheumatic fever for six months and his mother had totally devoted herself to nursing him. Mother, according to Mr Giles, was with him constantly. She got into bed with him.

     Later, it was discovered that before his illness Mr Giles had been preoccupied by a wish to invade his mother which came from two sources: first, he felt his mother abandoned him when his father became violent, as well as during her own depressed periods. The idea of getting inside her gratified an infantile wish to be close to her and, in this aggressive way, get from her all he had missed. Second, Mr Giles experienced his mother as pushing her fears and anxieties into him, leaving him confused and frightened. Mr Giles’s thoughts of getting into his mother also represented an identification with the aggressor: a wish to do to her what he felt she had done to him.

     Mr Giles suffered a physical intrusiveness from his father’s physical assaults and a psychological intrusiveness through his mother pushing her anxieties into him. Neither parent protected him from the other. The intrusiveness of both parents, as well as Mr Giles’s own wishes to get inside his mother, decisively influenced the way he experienced his confinement at the age of 8. Mr Giles told the therapist he was terrified by the pain which spread from the groin through his body until he could only move his head. In this vulnerable and painful state, Mr Giles had no resources with which to protect himself from his own wishes and the fears aroused by his mother’s seductiveness. Mr Giles’s wish to be close to his mother was gratified by her taking over care for his body.

     However, as we can see with Sylvia, wanting to be close leads to wishes to get inside the loved one, producing fears of suffocation and death. We speculate that the excitement was too incestuous to be represented as such. Due to this, the experience went through a developmental and representational regression, that is, from the possibility of an ordinary representation to a concrete presentation, in this case a suffocating closeness, which could be said to stand for the excitement. This presentation was experienced as a realised fact. The content of the presentation (i.e. the suffocating closeness) could in part be determined by the earlier experiences of incestuous excitement, helplessness and fear that were aroused by his mother.

     During therapy, Mr Giles acknowledged that he saw his mother as an incestuous figure who belonged to a violent, possessive and unpredictable husband. He admitted to fears of his father’s reprisals, which reinforced still further his anxieties about his physical defencelessness. Mr Giles defended against his helplessness when his father hit him or when faced with his mother’s projection by identifying with the aggressor, for example, when he bullied fellow students at school. His identification with an assaulting father reinforced his predilection to rely upon violence as a medium for defence.

     However, as we can see from his fantasies during the attack on Sylvia, it was his mother’s emotional intrusiveness and seductive nursing, which aroused primitive anxieties about annihilation and the need for a violent response. Mr Giles’s fears of bodily deterioration came from the conjunction of a real physical threat, the pain and immobilisation caused by the rheumatic fever, and an imagined physical threat, his fear of being suffocated aroused by his mother’s closeness and handling of his body. The illness itself deprived Mr Giles of an active motor response when he was faced with the threat of physical deterioration. Hence, Mr Giles was traumatised by being physically and mentally defenceless. He experienced physical intimacy with Sylvia as threatening to traumatise him in the same way his mother did, by stimulating his infantile wishes to invade her and the subsequent fears of suffocation. This triggered panic and self-preservative aggression.

     We believe that Mr Giles’s illness at age 8 is reported as significant by him because it repeated earlier traumatic states that had undermined his capacity to distinguish physical pain from anxieties associated with wish-fulfilling fantasies. We speculate that this early traumatisation involved a weakening of the weaving function which increased his vulnerability to threats towards his psychic cohesion. Early traumatisation could then be seen as leading to a vulnerable psychic fabric, reflected in the difficulty in distinguishing pain from anxiety and in threats of fragmentation.

     The illness itself deprived Mr Giles of an active motor response when he was faced with the threat of physical deterioration. A debilitating illness during latency could undermine one of the principal tasks of latency—the development of more sophisticated defences. This seems to have been the case for Mr Giles.

Enactment of violence in a session

Mr Giles’s sessions were often dominated by enactments of conflicts aroused by the transference or events outside the consulting room. At times, he abandoned language and expressed himself non-verbally. For instance, early on, Mr Giles presented himself as a scruffy, unkempt tramp who needed to be physically taken care of. He frequently fell asleep during sessions. His melancholic attitude would be defended against by intimidation, such as menacingly standing over his therapist while taking his coat off and throwing it over the spare chair, or spilling the contents of his pockets on to the therapist’s desk, or defiantly putting his feet up on the therapist’s desk, before collapsing back in his chair tired and dishevelled. Thus, Mr Giles’s enactment of sado-masochistic relating alternated between passive withdrawal and active intimidation.

     As one would expect in working with a patient who relies upon violence, Mr Giles’s violence broke through in sessions in response to disturbing manifestations of the transference. In the following vignette, one can see how an experience presented in a violent act finds a representational form through the work of the therapeutic couple.

     Mr Giles had missed his previous session because he had ‘forgotten’ that it was the start of a new session time. At the start of this session, Mr Giles was careful to protect the therapist from his anger about the failure to prevent a ‘misunderstanding’. He talked about a Hungarian chap who played with words, changing their meaning or using innuendo or words that might have an ambiguous meaning. The therapist took up the misunderstanding and Mr Giles’s missing his last session by saying that when Mr Giles feels there is a misunderstanding he really wonders what it is the therapist means to say. Can the therapist really be trusted to say what he means, to be available, to understand how angry Mr Giles is at the therapist for the missed session?

     Mr Giles denied blaming the therapist and said that he felt crowded and needed more space from his girlfriend. The therapist sensed steam building up as in a pressure cooker and felt anxious about his own safety. Mr Giles closed his eyes and hit the desk with his fist. The therapist was startled. The blow seemed to let out some of the steam. The therapist decided to stay silent.

     Mr Giles went on to talk about a Russian man who saw him with a drink at a party. Mr Giles was taking a drink to his girlfriend. The man thought he had caught Mr Giles breaking his abstinence, but clearly didn’t understand the situation. Mr Giles was pleased that he did not throw the drink at the man or punch him.

     The therapist said he was startled when Mr Giles hit the desk. He assumed Mr Giles was trying to get across what it feels like when the therapist does not understand what he needs, when the therapist does not understand that Mr Giles needed to be reminded about the new session time. It is difficult to get through, because Mr Giles also wants to protect the therapist, as he did earlier by closing his eyes and hitting the desk instead of the therapist. Mr Giles associated to an old friend who he is fond of because she says exactly what she means, regardless of what other people think or feel. He would like to express his aggression the way she does.

     One can see in this vignette that the metaphor of feeling crowded had been insufficient to contain Mr Giles’s anxieties associated with being intruded upon by the therapist. The sudden blow to the desk was a communication and a solution. At that moment, through an action, Mr Giles turned a passive experience into an active one. When Mr Giles hit the desk the therapist was startled and felt intruded upon. The therapist hadn’t expected it. Mr Giles startled the therapist, but the therapist did not respond immediately.

     Mr Giles’s action had also stopped what he experienced as the therapist’s intrusiveness, and that moment opened up the possibility of a change of tack. By hitting the desk, Mr Giles stopped the therapist’s interpretive line, which Mr Giles experienced as crowding him, and forced the therapist to think again. The therapist stayed with that moment and used it to reorient himself to Mr Giles. When Mr Giles said he felt crowded, the therapist began to see that Mr Giles felt the therapist was being intrusive. Returning to that moment with Mr Giles, the therapist used what had happened to himself to shape an interpretation that Mr Giles responded to by talking about an old friend who says exactly what she means. The interpretation was probably less important than telling Mr Giles that he could and did get through, that he could affect the therapist and that the therapist could use Mr Giles’s impact to try to understand him.

     This clinical vignette is an illustration of a ‘role response’ (Sandler, 1976) to pressure from the patient to ‘actualise’ the transference object. In this session, the analyst unwittingly enacted the parents’ (particularly the father’s) failure to perceive, process and use language to make contact with and understand their child. When this occurs with a patient whose primary objects were unable to take in the child’s projections of feelings and thoughts, there may be a breakdown in the effective use of symbols and metaphors to construct thoughts and feelings. The absence of an object into which the subject could project thoughts and feelings ‘increases the need for violent muscularity, to replace the failed projection. Were such symbolisation possible, projections would take place and therefore no violence would occur’ (Sohn, 1995, p. 574). Mr Giles momentarily abandoned symbolisation and resorted to action. However, the potential meaning of the act found representation in the ensuing communication as the therapist was able to stay with his countertransferential reaction; the iconic presentation was transformed into a representation in the therapeutic dialogue.


It is rare to find an account of a habitually violent man in regular psychoanalysis; Karl is one of the very few. Karl’s treatment is well documented in two papers written by Perelberg (1995, 1999).

     Karl came for psychoanalysis because of academic failure, but also due to problems with his peers and tutors, with whom he easily got into conflict. In the interviews a preoccupation with violent fantasies was apparent (he reported, for example, that he was convinced that his parents practised overt sado-masochistic sexuality), but not until some months into the analysis did the full extent of Karl’s violent behaviour become apparent. He had been, and was, involved in criminal activities, like selling stolen goods, but he had also resorted to overt violent behaviour, for example, fighting with broken bottles. The violent activities often began in a planned manner, but easily got out of hand.

     We cannot do justice to the complexity of Perelberg’s analysis or her multi-dimensioned understanding of Karl’s violence. We will only focus on one aspect of Karl’s violence which Perelberg saw as self-preservative, that is, as defensive measures to keep up a vital integrity (in the last reckoning identical to survival). In this respect, we found a common denominator in Karl’s violence and the violent behaviour of Mr Giles and Mr Wilson.

     Karl’s biological father had left Karl’s mother during the pregnancy, but, soon after the birth, the mother got married to another man who adopted Karl. Two sisters were born into the family, but Karl experienced a close as well as ‘special’ relation to his mother. The father was unpredictable, teasing and violent. During adolescence, Karl had sought help due to a preoccupation with body odour, disturbing sado-masochistic fantasies and suicidal ideation. When he came for psychoanalysis in his early twenties, Karl had come to a dead-end in his studies and peer relations, and had moved back into his parents’ house.

     Early in the analysis, it became apparent that sessions which included some meaningful dialogue were followed by a failure to attend; Karl overslept, often for many consecutive sessions. Somewhat later, it was clear that moments of contact led to a ‘blank space’ of which a dreamless sleep was one example. This reaction to felt closeness was partly substituted by (or alternated with) an activation of both criminality and violence. Karl became more elaborated in his violent fantasies and deeds, and it also became apparent that it was easier for Karl to attend when he had been involved in criminal activities. It thus seemed clear that the criminal and violent acts had become vicissitudes of something felt in the transference.

     The analyst suggested that Karl felt her to be intruding, and that the violence perhaps gave him an experience of rescuing some distance. Karl responded by talking about a gun he had acquired (in response to which the analyst started to wonder if she should interrupt the treatment). After some analytic work, Karl got rid of the weapon, but became overwhelmed by feeling lost, abandoned and depressed.

     Closeness to the analyst led to two alternative reactions. First, Karl could fall into a blankness, concretely denying the existence or value of the analyst by staying in a dreamless sleep. Second, Karl experienced closeness as suffocating; as equal to an ‘engulfment’. In order to create a space of his own, to distance himself from an engulfing mother, and to uphold a sense of integrity, Karl resorted to violence.

     This seemed to be the transferential matrix from which the violent acts ensued: whenever the safety in the transferential distance broke down, Karl had the experience of becoming engulfed and of losing his identity, and this was felt to be a concrete threat to his survival. In order to avoid the threat, Karl attacked.

     A vignette from a summer break: a community of scientologists had opened an office in Karl’s neighbourhood. Karl and his friends had become preoccupied with scientologist ideology and had concluded that it was ‘dangerous’. They sent threatening letters and made aggressive telephone calls. Finally, Karl went to their office and lost control, and the police were called. The arrival of the police was a relief to Karl.

     After some analytic work on these incidents, the analyst could see that when Karl missed her he became frightened of his dependent feelings, which were projected and led to Karl perceiving his analyst (represented by the scientologists) as dangerous. Karl resorted to violent fantasies to break free from what he experienced as an engulfing threat to his survival. Fighting the scientologists enabled Karl to shift from a passive victim mode to being an active perpetrator.

     A few months later Karl was pondering why he, again, had been unable to attend sessions. He wondered if oversleeping was connected to the fact that the one who woke him as a child was his mother: perhaps he did not want to wake up ‘to his mother’. He then came to a dream about the Incas. He had, in fact, been to South America in his youth. Around that time, he had been occupied by piranhas: he was amazed how they could eat an entire man. Karl’s mother gave him a stuffed piranha, and he just recently bought another one. The analyst thought this session shows a transferential fantasy. Waking up to, and for, the analyst leads to associations of dangerous, engulfing piranhas who might eat a whole man. Oversleeping protects Karl from such a dangerous object.

     We think that the case of Karl, like that of Mr Wilson and Mr Giles, illustrates Glasser’s view that in some individuals violence functions to restore by physical means a psychic equilibrium that has been threatened by anxieties associated with abandonment and/or engulfment. When feeling abandoned and dependent, Karl resorts to violence, and in the act towards the scientologists he presents his experience: the experience in the transference does not become represented, only presented. A few months later the same transference fantasy appears in a representation, that is, in the shape of engulfing, dangerous piranhas.

     Why do some individuals defend against primitive anxieties about survival by defence mechanisms that rely upon symbolisation and metaphor while others resort to violence? We believe the answer lies in a failure of metaphor to contain anxiety. With the breakdown of metaphorisation, the individual regresses to non-verbal modes of defence. When Mr Giles becomes violent and startles his therapist the experience of being intruded upon shows in the very act, like Karl who presents his transferential experience in the act towards the scientologists. We believe these examples also show how a presented experience can come to be represented due to psychotherapeutic or psychoanalytic work, and that this representation confirms the potential meaning of the very act: a presented latent meaning in the act becomes a represented one in the analytic work.

     We understand that the function of metaphor is to contain through symbolisation.


From this clinical material, we want to pick out only certain aspects. Our findings are partly speculative and call for further case studies.

     The experiential world is largely primarily shaped through bodily configurations. These configurations are originally iconic presentations—primitive mental formations equating a concrete act with an experience or an impulse—possibly turning into schema through which other experiential domains may be configured. This latter function is analogous to a metaphorical function in dreams and the transference where an unconscious wish or fantasy is given representational actualisation through a tangible medium.

     This configuration in bodily terms can be observed in Mr Giles’s experiences of invasion and engulfment. In his childhood, Mr Giles had been dominated by a wish to ‘invade’ his mother, partly in response to what he perceived as her ‘invasion’, and partly in response to anxiety about being abandoned by her. These ‘invasions’ were, naturally, not real but mental and experiential configurations of wishes, impulses, fears and relational experiences. However, the consequence of a successful invasion of mother aroused additional anxieties of engulfment. The invasion became identified with ‘engulfment’, namely, a fantasy of being eaten. What Mr Giles wanted from—and experienced in relation to—his mother was configured in terms of concrete actions of the body, and his anxiety was connected to a threat that these actions might be realised with Sylvia.

     When ill during latency, Mr Giles experienced his mother as invasive, while Mr Wilson felt his mother abandoned him. Nevertheless, both Mr Giles and Mr Wilson harboured intense, longstanding wishes to get back to their mothers which, in turn, created terrifying anxieties about their own extinction. Neither man could find refuge with his envious, violent, unapproachable father. Consequently, Mr Giles and Mr Wilson did not develop effective, socially acceptable defences.

     Mr Wilson shared Mr Giles’s impulse to invade his girlfriend, but his wish was triggered by another kind of anxiety. Mr Giles began to hit his girlfriend Sylvia because he was terrified of being trapped inside her and suffocating to death, while Mr Wilson identified his girlfriend with a mother who he believed abandoned him. Mr Wilson’s anxiety grew out of the fear he would never reach his girlfriend and so, as he put it, ‘be left to die’. This ‘being left to die’—an experience of objectless desolation and catastrophic emptiness—was configured as ‘blackness’ that Mr Wilson tried to escape by punching, and through punching make contact with ‘someone, something, anything’.

     Karl felt he had a close and ‘special’ relationship with his mother, which excluded his father and sisters. We understood that the price he had to pay for this closeness was a conviction that she would not allow him any independence (including a sexuality of his own). Closeness was, as with Mr Giles, configured as an experience of ‘engulfment’. In the transference, Karl felt the threat of engulfment as a concrete risk and he presented this, first, in a concrete violent act. Subsequently, the same experience was represented in a fantasy of piranhas eating. Likewise, in the vignette from Mr Giles’s therapy, a transference presentation was seen to find representation in the therapeutic dialogue.

     In spite of their differences, Mr Giles and Mr Wilson share an experience of traumatisation. Mr Giles was traumatised by being trapped in bed, becoming ‘invaded’ by an all too stimulating mother and a threatening illness (echoing earlier experiences, fears and wishes of ‘intrusion’). Mr Wilson was traumatised by the fear of being left on his own in hospital to die. We do not know if Karl had had similar experiences, but in the clinical case report one can perceive how a passive position led to a threat of losing both identity and integrity.

     Mr Giles and Mr Wilson’s fear of bodily deterioration had been decisively influenced by the conjunction of a physical threat—i.e. bodily illness during latency—and a psychical threat (the fear of being invaded by a seductive mother or left to die by an abandoning one). Their mothers’ responses at a time when both men believed their bodies to be in danger reinforced a link between primitive annihilation anxieties of engulfment or abandonment and physical survival that was revived later in adult heterosexual relationships. Being unable to cope with an emotional situation was in this way strongly connected to a fear of bodily deterioration, and the latter fear became a bearer of the former one. The experiences during their illnesses thus came to reinforce both men’s proneness to experience emotional threats as concrete ones.

     Something analogous to this can be seen in Karl’s conviction that his parents practised sado-masochistic sexuality. True or not, Karl was convinced that sexuality and closeness implied violence, which became clear in the transference.


The child might use schemes of bodily sensations and functions through which he can categorise, and thus give meaning to, external and internal worlds. In this paper, we have compared this representational function to ordinary metaphorical functioning in which one domain of the world is given form through another, better-known, one. In the cases of violence referred to above there seems to be a regression from this level of functioning to the one preceding it, namely, the level where representations are presentations in action. These can be viewed as concrete forms or figures analogous to concretised metaphors where a representative link has been replaced by a concrete equation.

     We have outlined a model in which a basic representational matrix, also conceptualisable as the body ego, is the condition for a metaphorisation of the mind. We suggest that the violent acts referred to above can be viewed from the perspective of this model. When the body-ego matrix is laid bare in a destruction of the weaving function, the subject senses this as a threat to his survival as a narcissistic whole. In order to avoid fragmentation, the subject gives form to the experience at the level of concrete presentation, which is the violent act.

     There seem to be certain parallels between concretised metaphors and the violent act. The perpetrator might use metaphors in an ordinary way, benefiting from the psychic work inherent in the metaphor’s interconnecting activity. The loss of differentiation between subject and object leads, however, to a collapse in the weaving function. Ordinary metaphorisation becomes impossible, but the pre-formed metaphor (e.g. ‘her nagging naïve arguments. . .they’ll be pushed up my arse’) is used in an attempt to maintain a psychic cohesion. Even though the body scheme cannot be used as a representative configuration through which something else is viewed, the scheme itself is used as a concrete presentation of the experience in question; the violent act comes, so to speak, to represent the actuality of the situation. The violent act is thus—for these subjects—a concrete configuration giving form to an experience, but this configuration does not lead to what can be called a psychological elaboration that characterises ordinary metaphors. The act thus remains compulsive and desolate.

     In conclusion, we would like to make five points:

  1. Mr Wilson, Mr Giles and Karl configured their inner situation in bodily terms, that is, as something happening to or with their bodies.
  2. All these men sensed inner dangers that were represented in and through their bodies.
  3. These configurations of their internal states were not sensed as ordinary metaphors. Mr Giles and Karl did not feel like being trapped, and Mr Wilson did not feel like being left in the dark: Mr Giles and Karl felt they were engulfed and Mr Wilson thought he was in a life-endangering, isolating blackness. It is worth noting that all three men were, in fact, articulate, and used metaphors in an appropriate way during ordinary conversation. However, metaphors failed to function in an ordinary way when Mr Giles and Mr Wilson felt themselves to be in certain forms of danger. In these situations what for some people could have been ordinary and helpfully shaping metaphors became—in their functions—iconic presentations of real, bodily dangers. We speculate that all three men experienced a developmental and representational regression due to threats to narcissistic wholeness. This regression goes from a metaphoric function to an iconic one where the body scheme no longer mediates a picture any more, but is reality.
  4. In this way, Mr Wilson, Mr Giles and Karl configured a threatening inner situation in concrete metaphors of dangers to the body. However, it is possible that these dangers were experienced even more concretely than can be inferred from what these men afterwards relate in their psychotherapeutic or psychoanalytic sessions. It might be that the dangers were even less psychically represented at the related moments, and to a greater extent represented only in the concrete acts (i.e. in the violent acts). We have the sense that all three men possibly reconstruct their experiences in more elaborated terms than what occurred during the assaults. If so, the inner situation might have been manifested in the very act rather than in an experience of concretised metaphors. This can hardly be decided, but one can infer that the attacks in themselves are highly concrete configurations giving perhaps—in these situations—the only possible form to the dangers in question. Following this, the assault would be the ‘poem’ of the body in a situation in which the threats are not even experienced as such; in this way one can see the violent act as the most concrete metaphorisation through the body, like the ‘archaic hysterias’ described by McDougall (1989) and Campbell (2000).
  5. These concrete configurations nevertheless have a rescuing function, as they seem to save the men from a deteriorating psychic situation. The ‘real’ threats—which in themselves are psychotic productions—trigger a concrete response (i.e. the violent act) as an attempt at rescue from fragmentation. We suggest that the concrete metaphorisation of the assault keeps up a cohesive mental configuration, albeit a concrete one. This interpretation gets support from Tuovinen’s (1973) investigations of homicides which were often found to be self-preservative attempts both to represent danger and to defend against it in order to avoid a perceived threat of psychotic fragmentation.

     A stable psychic fabric makes ordinary general metaphorisation possible. When this metaphorisation functions, it binds tension in psychic work, and this is a way to translate the fact that representations come to contain experience.

     However, there are instances in which the conditions for ordinary metaphorisation break down, as when differentiation between subject and object fails, and these lead to psycho-economic crises as well as a failing of reality testing. In order to prevent a psychotic fragmentation a subject might use a pre-formed metaphoric configuration which keeps up a cohesion, but the vitality of a functioning metaphorisation of the mind is unattainable—and this includes an inability to contain the threatening upheaval. In order to defend against the threatening fragmentation, the perpetrator represents—and defends against—the danger through a concrete act, such as a violent assault.

     One can observe how the body ego in these violent acts is ‘reactivated’ in two respects. On the one hand, the subject uses the body-ego matrix to configure (in an iconic way) the instinctual threat, and, on the other hand, he uses the motoric means of the body ego to defend against the threat. These ‘poems of the body’ remain, however, empty and desolate, as they do not contain symbolic representations of vital metaphors which can facilitate integration and growth. This, however, may become possible in a psychotherapeutic or psychoanalytic treatment where the acts can find a representation.



End Notes

1 Corresponding author.

2 The Portman Clinic is a British National Health Service outpatient facility that offers assessment, consultation on management and psychoanalytic psychotherapy to patients with delinquent behaviour or suffering from a sexual deviation.

3 The violent act has been the focus of a research project carried out by a group of colleagues who are treating violent individuals at the Portman Clinic including, Mervin Glasser, (Chairman), D. Campbell, S. Dermen, R. Doctor, C. Fishman, U. Gurisik, R. Hale, C. Lucas, B. Krikler, D. McLean, B. O’Neill, M. Parsons, A. Pieczanski, N. Temple and A. Zachary. See Glasser (1998) for a further description of methodology and hypotheses developed during this research project.



Arlow J (1969a). Fantasy, memory, and reality testing. Psychoanal Q 38:28-51.

Arlow J (1969b). Unconscious fantasy and disturbances of conscious experience. Psychoanal Q 38:1-27.

Arlow J (1979). Metaphor and the psychoanalytic situation. Psychoanal Q 48:363-85.

Beardsley M (1958). Aesthetics: Problems in the philosophy of criticism. New York: Harcourt, Brace and World.

Beardsley M (1962). The metaphorical twist. Phil Phenomenological Res 22:293-307.

Bird B (1972). Notes on transference: Universal phenomenon and hardest part of analysis. J Am Psychoanal Assoc 20:267-301.

Black M (1962). Models and metaphors. Ithaca, NY: Cornell Univ. Press.

Black M (1979). More about metaphor. In: Ortony A, editor. Metaphor and thought. Cambridge: Cambridge Univ. Press.

Borbely AF (1998). A psychoanalytic concept of metaphor. Int J Psychoanal 79:923-36.

Campbell D (2000). Pain and betrayal. Paper presented before the Finnish Psychoanalytical Society, September.

Corradi Fiumara G (1995). The metaphoric process: Connections between language and life. London: Routledge.

Enckell H (2001). Psychic reality and psychoanalysis: A look through the theory of metaphor. Psychoanal Contemp Thought 24:231-52.

Fonagy P, Target M (1995). Understanding the violent patient: The use of the body and the role of the father. Int J Psychoanal 76:487-501.

Freud S (1900). The interpretation of dreams. SE 4-5.

Freud S (1905). Fragment of an analysis of a case of hysteria. SE 7.

Freud S (1914). Remembering, repeating and working-through. SE 12.

Freud S (1920). Beyond the pleasure principle. SE 18.

Freud S (1921). Group psychology and the analysis of the ego. SE 18.

Freud S (1925). Negation. SE 19.

Gaddini E (1982). Early defensive fantasies and the psychoanalytical process. Int J Psychoanal 63:379-88.

Gammelgaard J (1998). Metaphors of listening. Scand Psychoanal Rev 21:151-67.

Glasser M (1998). On violence: A preliminary communication. Int J Psychoanal 79:887-902.

Green A (1986). On private madness. London: Hogarth.

Kitayama O (1987). Metaphorization—Making terms. Int J Psychoanal 68:499-509.

Klein M (1952). The origins of transference. Int J Psychoanal 33:433-8.

Lakoff G, Johnson M (1980). Metaphors we live by. Chicago, IL: Univ. of Chicago Press.

Laplanche J, Pontalis J-B (1973). The language of psychoanalysis. New York, NY: Norton.

Lehtonen J (1997). On the origins of the body ego and its implications for psychotic vulnerablility. In: Volkan V, Akhtar S, editors. The seed of madness. Madison, CT: Int. Univ. Press.

Lewin B (1946). Sleep, the mouth, and the dream screen. Psychoanal Q 15:419-34.

Lewin B (1948). Inferences from the dream screen. Int J Psychoanal 29:224-31.

Lewin B (1971). Metaphor, mind and manikin. Psychoanal Q 40:6-39.

Loewald H (1975). Psychoanalysis as an art and the fantasy character of the psychoanalytic situation. J Am Psychoanal Assoc 23:277-99.

McDougall J (1989). Theaters of the body. New York, NY: Norton.

Melnick BA (1997). Metaphor and the theory of libidinal development. Int J Psychoanal 78:997-1015.

Ogden TH (1997). Reverie and metaphor: some thoughts on how I work as a psychoanalyst. Int J Psychoanal 78:719-32.

Opatow B (1997). The real unconscious: Psychoanalysis as a theory of consciousness. J Am Psychoanal Assoc 45:865-90.

Perelberg RJ (1995). A core phantasy in violence. Int J Psychoanal 76:1215-31.

Perelberg RJ (1999). The interplay between identifications and identity in the analysis of a violent young man: Issues of technique. Int J Psychoanal 80:31-45.

Proust M (1934). Remembrance of things past. New York, NY: Random House.

Richards IA (1936). The philosophy of rhetoric. Oxford: Oxford Univ. Press.

Ricoeur P (1970). Freud and philosophy. New Haven, CT: Yale Univ. Press.

Ricoeur P (1978). The rule of metaphor. London: Routledge and Kegan Paul.

Rizzuto A-M (2001). Metaphors of a bodily mind. J Am Psychoanal Assoc 49:535-68.

Salonen S (1979). On the metapsychology of schizophrenia. Int J Psychoanal 60:73-81.

Salonen S (1989). The restitution of primary identification in psychoanalysis. Scand Psychoanal Rev 12:102-15.

Sandler J (1976). Countertransference and role responsiveness. Int Rev Psychoanal 3:43-7.

Sandler J, Sandler A-M (1983). The ‘second censorship’, the ‘three box model’ and some technical implications. Int J Psychoanal 64:413-25.

Sandler J, Sandler A-M (1984). The past unconscious, the present unconscious and interpretation of the transference. Psychoanal Inq 4:367-99.

Sandler J, Sandler A-M (1998). Internal objects revisited. London: Karnac Books.

Searles H (1962). The differentiation between concrete and metaphorical thinking in the recovering schizophrenic patient. J Am Psychoanal Assoc 10:22-49.

Segal H (1957) Notes on symbol formation. Int J Psychoanal 38:391-45.

Segal H (1978) On symbolism. Int J Psychoanal 59:315-9.

Shahly V (1987). Eating her words: Food metaphor as transitional symptom in the recovery of a bulimic patient. Psychoanal Study Child 42:403-21.

Sharpe EF (1940). Psycho-physical problems revealed in language: An examination of metaphor. Int J Psychoanal 21:201-13.

Shengold L (1981). Insight as metaphor. Psychoanal Study Child 36:289-306.

Smith H (1992). Screen language and developmental metaphor. Ann Psychoanal 20:95-113.

Sohn L (1995). Unprovoked assaults: Making sense of apparently random violence. Int J Psychoanal 76:565-75.

Stern D (1985). The interpersonal world of the infant. New York, NY: Basic Books.

Szajnberg N (1985). Übertragung, metaphor and transference in psychoanalytic psychotherapy. Int J Psychoanal Psychother 11:53-70.

Székely L (1962). Symposium: The psycho-analytic study of thinking. Int J Psychoanal 43:297-305.

Tuovinen M (1973). Crime as an Attempt at Intrapsychic Adaptation. Acta Universitatis Ouluensis, Series D, Medica No. 2, Psychiatrica No. 1. Oulu: University of Oulu.

Walker N (1968). Crime and punishment in Britain. Edinburgh: Edinburgh Univ. Press.

Williams A (1960). A psycho-analytic approach to the treatment of the murderer. Int J Psychoanal 41:532-9.

To cite this article, use this bibliographical entry: Donald Campbell "Metaphor and Psychoanalysis: Metaphor and the Violent Act". PSYART: A Hyperlink Journal for the Psychological Study of the Arts. December 15, 2009. Available August 12, 2006 [or whatever date you accessed the article].
Received: June 1, 2006, Published: August 12, 2006. Copyright © 2006 Donald Campbell